We had some fun with one of our sites, building out the email funnel in the 7 or 8 months since acquisition.
Interested in hearing this story?
Well, we’re excited to tell you.
Let’s dive right in.
The Origin Story
We always look for quality sites with strong traffic growth, quality backlink profile, and stable earnings history.
We prefer “stable” over a “fixer upper” type sites.
This particular site ticked all of those boxes, but it also had a growing email list and large traffic volume, which got us excited about reducing its reliance on SEO as a single traffic channel.
The seller had created an ebook and launched it to their list two months earlier, but did nothing with it since.
Wanting to diversify our revenue streams away from just affiliate commissions and display ads, we decided to build a sales funnel for the ebook.
To give you some context, ebook sales made up <2% of revenue, but now it has increased to around 10%.
Here’s what it looked like before we did anything.
- The site was generating pretty good traffic – over 1MM monthly pageviews
- It had an ebook for sale, but conversions were low
- It had an email list of around 13,000 that was lukewarm (at best) from a generic newsletter opt-in embedded in the footer with (as you guessed) terrible conversion rates
We imported all subscribers to ConvertKit in December 2019, but didn’t really do much to grow it until March 2020.
Here’s a graph showing the Net New Subscribers from February 26 to March 25, 2020. We had an average of 10 new opt-ins per day.
(For simplicity’s sake, I’m only going to talk about New Subscribers instead of Net New Subscribers from here on out.)
Part 1: Getting Started
There are 3 parts to this story, but let me paint you the overall picture first.
Our email list grew from 13,548 (March 20) to 20,902 (June 23) – that’s 7354 new subscribers or 54.3% increase over 3 short months!!
Yes, we dipped a bit (see 2) but came back stronger (see 3).
Keep reading to find out why that happened and how you can avoid the same mistake.
Let’s go back to the beginning.
March 26 was when we turned on all the new opt-ins and email automation.
Immediately, we saw it spike up.
Here’s the breakdown:
- March 25 (before we turned it on) – 11 new subscribers
- March 26 (we turned it on) – 37 new subs
- March 27 – 52 new subs
- March 28 – 71 new subs
Wow, from around 10 to 70 opt-ins daily – that’s 7x growth in just 3 days!
When things stabilized a few weeks later, we still had a daily opt-in average in the low 60s.
What did we do?
Here’s what we did before the funnel was turned on:
- Designed and created Lead Magnets – we created specific magnets from the ebook and put it on relevant pages. We had 13 in total.
- Designed and created Opt-In forms – we created a lightbox as well as in-content for each lead magnet. That’s 13 opt-ins x 2 types = 26 total.
- Wrote a 3-email welcome series
- Configured ConvertKit – this included setting up tags, forms, automations…etc
- Connected ConvertKit with ThriveLead
- Configured ThriveLead – the right display & pop up settings
- Tested everything ourselves multiple times
- Set up a Facebook group to start a community
Here’s what we did after the funnel was on:
- Created more lead magnets and corresponding opt-ins
- Started to split-test the old & new opt-ins – we alternated between just copy vs image/color
- Created 3 more ebooks so we have other offers to promote
- Combined all 4 ebooks together as a bundle
- Wrote more email series to take subscribers to subsequent stages of the funnel
- Created more tags, automations & forms in ConvertKit
- Grew the Facebook group from 0 to 750
- Scheduled the next 6 months’ worth of email campaigns promoting the ebooks
Part 2: The Dip
It happened suddenly.
We couldn’t understand why initially – things were going as it should for the last 2 months. Then overnight (May 19)…devastation hit.
The number of daily opt-ins dropped. Big time.
We went from a conservative daily average of 130 new subscribers to an optimistic 55. That’s a 57.7% drop!
It had me questioning and triple checking everything.
Were the numbers and stats in ThriveLead and ConvertKit accurate? Did we configure things correctly? Is there a pattern I’m not seeing?
What did we do?
Nothing…
The hardest part is to wait until the test is complete.
Best practices say to NOT touch it.
We repeat – DO NOT TOUCH IT.
…Not until enough people have gone through to either confirm or invalidate your hypothesis. It would be too premature to end the experiment otherwise.
Why?
Because the population size is too small to base an accurate conclusion off of.
The rule of thumb we usually stick to is 1000 people or 2 weeks’ time.
We waited 2 weeks – May 19 to June 1.
That’s 14 days of watching our numbers dwindle down – a bit disheartening to say the least.
In total, we had about 750~ people or so go through the top funnel. we felt it was enough data gathered to make a move.
And this is how we sprung back to life.
Part 3: The Bounce Back
On June 2, we regained our footing and our stats went above and beyond what we had previously.
Below we detail exactly what we did to correct this situation.
What did we do?
As mentioned, we questioned and triple-checked everything, including if:
- Configurations on all software tools were set up correctly
- There were any outliners or patterns that data can show me
- All moving parts of the top funnel were integrated and working properly
- Any of our recent actions had secondary impact we haven’t thought – we had to retrace our digital steps
We had a hunch what the culprit was, but we were confused.
Why?
Because it wasn’t the first time we did this, so why did this time have such a big effect?
All we can say is…we simply reached critical mass.
The culprit was the opt-ins.
The issue can be broken down into 2 factors.
1) Too many opt-ins
So, we went back to basics.
We pulled up Google Analytics and looked at the top 25 pages with the most traffic.
We deactivated all the opt-ins that weren’t specific to these pages (besides the general site-wide pop-up, of course).
2) Running too many split-tests at the same time
Within this new group of active opt-ins, we looked at the data of each test and did 1 of 2 things.
We either a) kept the test going due to lack of data, or b) chose the (obvious) winner and stopped the test.
In addition, display settings were adjusted accordingly and new, more specific lead magnets were created for top pages we didn’t account for before.
Here are a few examples of how some opt-ins performed:
When we went from 28 active opt-in forms to 17, our optimistic 55 daily new subscribers increased to a conservative average of 150. That’s at least a 172% jump!
When in doubt, just simplify.
(Bonus) Part 4: The Epilogue
Since we were on the up-and-up, now we can put our attention on the rest of the funnel.
We noticed that the ebook bundle wasn’t converting as well as we hoped, and decided to create a more desirable product.
We surveyed our audience.
A single email about a questionnaire yielded a respectable 520 completed responses.
What an active and engaged list!
After slicing and dicing the data, 3-4 broad themes were identified.
Based on data, we decided on the details of this product: what it entails, format, delivery method, price points…etc.
The people who said they would buy it are telling us exactly what they want after all – we better listen.
We threw up a pre-order landing page, notified our awesome email list and…crickets!
We don’t get it…they told us they wanted this, so why aren’t they buying?!
A few more follow-up emails were sent.
In the end, only 2 people bought.
Unfortunately, the minimum number we wanted to see was 5.
Refunds issued and back to the drawing board.
What will we do next?
There are many things we haven’t tried yet that we look forward to. Here are 5 off the top of our head:
- Create upsells, downsells and cross-sells
- Partner with similar sites or physical stores
- Paid newsletter, paid community or paid online support group
- Collaborate with influencers
- White label our own physical products
In terms of priority, we’d probably start with the one that takes the least amount of effort and time.
5 Lessons to Take Away
We boiled down what we learned on this rollercoaster ride into 5 bit-sized lessons for you.
- Tools are important
- We used ConvertKit for email platform & ThriveLeads for opt-in — they served us well but we’re thinking of moving to different tools. Currently still researching, but leaning towards ActiveCampaign for email. Unsure for opt-in software yet. (If you have suggestions, please comment below.)
- Simplify, simplify, simplify!
- When in doubt, the simplest option is best. When things get too messy, it’s time to clean up. When things get too complicated, it’s time to bring it back to basics.
- Don’t believe everything your audience tells you
- Just remember that most people know what they don’t want instead of what they want.
- Just to give you a polarizing point of view: the late Steve Jobs once said “People don’t know what they want until you show it to them. That’s why I never rely on market research. Our task is to read things that are not yet on the page.”
- Keep testing
- Every business is different, so what may work for you may not work for me. That’s why it’s vital to keep brainstorming new ideas, thinking outside of the box and applying frameworks creatively.
- Patience is a virtue
- Sometimes speed isn’t everything. Sometimes the best course of action is to wait it out.
And, that’s it!
Hope you enjoyed this case study and are able to implement the takeaways.
Let us know how we helped you, or if you have any suggestions for us to try!
